Re: Is this a good definition for clustering factor
- From: Niall Litchfield <niall.litchfield@xxxxxxxxx>
- To: breitliw@xxxxxxxxxxxxx
- Date: Thu, 17 Feb 2005 11:18:55 +0000
On Wed, 16 Feb 2005 18:18:35 -0700, Wolfgang Breitling
<breitliw@xxxxxxxxxxxxx> wrote:
> Why invent another wheel? What is wrong with Oracle's definition?
> Oracle 9i (9.2) Database Reference:
> Indicates the amount of order of the rows in the table based on the values=
> =20
> of the index.
> - If the value is near the number of blocks, then the table is very well=20
> ordered. In this case, the index entries in a single leaf block tend to=20
> point to rows in the same data blocks.
> - If the value is near the number of rows, then the table is veryrandomly=20
> ordered. In this case, it is unlikely that index entries in the same leaf=20
> block point to rows in the same data blocks.
I like the tuning guide example. I don't like the descriptions above.
In particular it seems to me that the first sentence is very poor
english 'amount of order' is just an ugly phrase. I think that I would
have used something like 'indicates how well the physical ordering of
the rows in the table matches the order of the index'.
In addition both of the sentences " If the value is near the number of
blocks, then the table is very well ordered." and especially "If the
value is near the number of rows, then the table is very randomly
ordered." would benefit from the addition of "by the keys of the
selected index".The ordering of the table only makes sense in relation
to the index being used to access it.
--
Niall Litchfield
Oracle DBA
http://www.niall.litchfield.dial.pipex.com
--
//www.freelists.org/webpage/oracle-l
Other related posts:
